1. Field of the Invention
This invention relates to measurement of fluid flow.
2. Prior Art
Various apparatus for the measurement of fluid velocity are known. First, a system for measuring the relative velocity between a body and a contiguous fluid includes an impeller or rotor which is rotatably pivoted on the body immersed in the fluid. Flow of the fluid relative to the body causes the impeller or rotor to rotate with an angular velocity dependent on the relative velocity between the body and the fluid. This rotation may operate, by electrical or mechanical means, an indicator which is suitably calibrated to indicate relative velocity. Disadvantages of such an arrangement include that the angular velocity of the impeller or rotor is not accurately proportional to the relative velocity between the fluid and the body.
An apparatus related to the previously discussed impeller rotor is a vane-like obstruction in the fluid path. The amount of deflection of the vane is proportional to the fluid velocity. Deflection of the vane can be sensed by any of a variety of means such as movement of a slider along a resistance thereby varying the resistance proportional to vane deflection.
Also known are hot wire fluid flow sensors wherein the motion of the fluid past a heated wire cools the wire. The amount of cooling is dependent upon the amount of fluid flowing past the wire. The amount of cooling of the wire can be determined by a resistance change which can be measured in any number of ways. Thus, a hot wire fluid flow sensor can measure the mass of the fluid flowing past the sensor. It is advantageous to know the mass of the air entering an internal combustion engine when establishing an air fuel ratio. Disadvantages of such an apparatus include that various impurities in the fluid may be deposited on the wire and cause an erroneous indication of fluid mass flow.
Another method of determining fluid velocity uses an object placed in the fluid flow to create a fluctuating flow of vortices in the wake of the object located in the fluid, i.e., Karman vortices. Alternatively, vanes can be placed in the fluid flow to induce a swirl having a cork screw shape. The rate of passage of the ridges and troughs of the cork screw are an indication of fluid velocity.
It is known that the Karman vortices tend to be formed on the downstream side of the object at regular intervals and alternately, first behind one side or edge of the object and then behind the opposite side or edge. The vortices detach themselves from the object in two nearly parallel rows and are carried downstream at a velocity substantially proportional to, but somewhat less than, the relative velocity of the fluid. That is, the vortices have a slip velocity which tends to bear a constant ratio to the velocity of the fluid. It is also well known that in each row the vortices are formed at a distance apart which tends to bear a constant ratio to the size of the object but substantially independent of the relative velocity of the fluid. From these considerations it has been shown that the frequency at which vortices are formed behind either side or edge of a cylinder tends to be proportional to the relative velocity of the fluid and inversely proportional to the size of the cylinder, but to depend substantially upon no other factor or parameter.
The phenomenon of alternate vortex formation in two rows in this way occurs frequently in the natural world, and is known to be the cause of, for example the "singing" of telegraph wires in the wind, the "sighing" and "roaring" of wind in trees, and the "whistling" of wind through tall grasses.
However, reliable, accurate and inexpensive measurement of these vortices to determine fluid flow has presented a problem. For example, U.S. Pat. No. 3,116,639 issued to W. G. Bird teaches positioning a vane-like element mounted for oscillatory movement about an axis extending at right angles to the direction of flow of the fluid and positioned downstream of the vortex generator. The frequency of oscillation of the vane-like element is measured and is related to the frequency with which vortices pass the element. The response time of such a system is, of course, related to the mechanical response of the system. That is, the vane-like element may not be able to respond to the changes in velocity. Further, the strength of the vortices must be sufficiently strong to cause movement of the vane-like element.
Another method and apparatus for determining the passage of a vortex is taught in U.S. Pat. No. 3,680,375 issued to R. D. Joy et al. A sonic signal transmitting transducer is located on one side of the wake of vortices and a sonic signal receiving transducer is located on the other side of the wake. The transmitting transducer signal is modulated by the Karman vortices and received by the receiving transducer. An electronic signal detector is connected to the signal receiving transducer for detecting the modulations created by the Karman vortices. Here again, the vortices must be of sufficient magnitude to cause modulation of the sonic signal. Further, modulation of a sonic signal may occur due to disturbances in the fluid flow other than vortices thus causing erroneous signals. That is, the turbulence or random velocity fluctuations within the fluid stream are similar to the vortices generated by the vortex generator. These turbulent fluctuations result from obstacles that the fluid previously encountered as well as the turbulence generated at the wall boundary layer. Recent studies suggest that the turbulence at the wall is formed by random turbulence bursts. Minimizing these turbulent bursts will result in a lower noise and thus an increased signal to noise ratio. This invention overcomes some of these problems by reducing the turbulence generated at the wall by controlling these turbulent bursts.
Also known are studies describing the drag and heat transfer characteristics of longitudinally ribbed surfaces. For example, in "Drag and Heat-Transfer Characteristics of Small Longitudinally Ribbed Surfaces", by M. J. Walsh and L. M. Weinstein, AIAA Journal, Vol. 17, No. 7, pp. 770 and 771, 1978, longitudinally ribbed surfaces are examined with the purpose of confining turbulent bursts to regions of small transverse extent, thus altering the local turbulence production. Rectangular, triangular and razor blade rib geometrics were examined. "Drag Characteristics of V-Groove and Transverse Curvature Riblets" by Michael J. Walsh, Viscous Flow Drag Reduction, pp. 168-184, Vol. 72 of Progress in Astronautics and Aeronautics, published by AIAA, 1980, also examines the drag characteristics of longitudinally ribbed surfaces where the rib dimensions are the order of the initial turbulent wall burst region near the wall.